Face morphing attacks represent a serious threat to biometric face recognition systems deployed in real-world operational environments, such as Automated Border Control (ABC) gates. Therefore, it is fundamental to develop accurate systems to counter this attack, i.e. Morphing Attack Detection (MAD) systems.
Automated Border Control (ABC)
gates in airports
While most existing Morphing Attack Detection (MAD) approaches focus on single images (S-MAD) or image pairs (D-MAD), operational systems routinely acquire short video sequences, which are largely underexploited by current methods.
This competition is motivated by the need to bridge this gap between research and practice by fostering the development of Video-based Morphing Attack Detection (V-MAD) approaches that explicitly leverage temporal information, multiple frames, and varying image quality.
The goal of the competition is to advance the state of the art in morphing attack detection by encouraging methods that can effectively operate on video data and improve the security and reliability of face recognition technologies in operational settings.
Notably, the competition introduces a new task definition (V-MAD) and new benchmarks, not yet available in the literature. Further details on V-MAD can be found in the original paper
Borghi, G., Franco, A., Di Domenico, N., Ferrara, M., Maltoni, D.
V-mad: Video-based morphing attack detection in operational scenarios.
In IEEE International Joint Conference on Biometrics (IJCB), 2024
In operational scenarios, V-MAD represents a viable paradigm when a single probe document image is compared with an input sequence to detect whether the image is morphed or not.
All participating teams will be required to submit their solution on the free and open-source Codabench platform, which supports the upload of models, their evaluation, and the display of results.
The authors of the best-performing models will be invited to contribute to a final paper of the competition.
Ths challenge is organized in conjunction with the
IEEE International Joint Conference on Biometrics (IJCB 2026)
1 - 4 September 2026, Rome, Italy